GP‐FMLNet:A feature matrix learning network enhanced by glyph and phonetic information for Chinese sentiment analysis

Sentiment analysis is a fine‐grained analysis task that aims to identify the sentiment polarity of a specified sentence.Existing methods in Chinese sentiment analysis tasks only consider sentiment features from a single pole and scale and thus cannot fully exploit and utilise sentiment feature information,making their performance less than ideal.To resolve the problem,the authors propose a new method,GP‐FMLNet,that integrates both glyph and phonetic information and design a novel feature matrix learning process for phonetic features with which to model words that have the same pinyin information but different glyph information.Our method solves the problem of misspelling words influencing sentiment polarity prediction results.Specifically,the authors iteratively mine character,glyph,and pinyin features from the input comments sentences.Then,the authors use soft attention and matrix compound modules to model the phonetic features,which empowers their model to keep on zeroing in on the dynamic‐setting words in various positions and to dispense with the impacts of the deceptive‐setting ones.Ex-periments on six public datasets prove that the proposed model fully utilises the glyph and phonetic information and improves on the performance of existing Chinese senti-ment analysis algorithms.

On the Linguistic Features of Comments in Micro-blogs from Different Social Status

The study makes a contrastive analysis of the differences in the comments of people from different social status in their Mini-blogs.It aims to find the manifestations of those differences by analyzing the underlying causes and provide constructive suggestions on how to build a more harmonious public platform,which is of its great significance.

Assessment of Sentiment Analysis Using Information Gain Based Feature Selection Approach

Sentiment analysis is the process of determining the intention or emotion behind an article.The subjective information from the context is analyzed by the sentimental analysis of the people’s opinion.The data that is analyzed quantifies the reactions or sentiments and reveals the information’s contextual polarity.In social behavior,sentiment can be thought of as a latent variable.Measuring and comprehending this behavior could help us to better understand the social issues.Because sentiments are domain specific,sentimental analysis in a specific context is critical in any real-world scenario.Textual sentiment analysis is done in sentence,document level and feature levels.This work introduces a new Information Gain based Feature Selection(IGbFS)algorithm for selecting highly correlated features eliminating irrelevant and redundant ones.Extensive textual sentiment analysis on sentence,document and feature levels are performed by exploiting the proposed Information Gain based Feature Selection algorithm.The analysis is done based on the datasets from Cornell and Kaggle repositories.When compared to existing baseline classifiers,the suggested Information Gain based classifier resulted in an increased accuracy of 96%for document,97.4%for sentence and 98.5%for feature levels respectively.Also,the proposed method is tested with IMDB,Yelp 2013 and Yelp 2014 datasets.Experimental results for these high dimensional datasets give increased accuracy of 95%,96%and 98%for the proposed Information Gain based classifier for document,sentence and feature levels respectively compared to existing baseline classifiers.

SA-MSVM:Hybrid Heuristic Algorithm-based Feature Selection for Sentiment Analysis in Twitter

One of the drastically growing and emerging research areas used in most information technology industries is Bigdata analytics.Bigdata is created from social websites like Facebook,WhatsApp,Twitter,etc.Opinions about products,persons,initiatives,political issues,research achievements,and entertainment are discussed on social websites.The unique data analytics method cannot be applied to various social websites since the data formats are different.Several approaches,techniques,and tools have been used for big data analytics,opinion mining,or sentiment analysis,but the accuracy is yet to be improved.The proposed work is motivated to do sentiment analysis on Twitter data for cloth products using Simulated Annealing incorporated with the Multiclass Support Vector Machine(SA-MSVM)approach.SA-MSVM is a hybrid heuristic approach for selecting and classifying text-based sentimental words following the Natural Language Processing(NLP)process applied on tweets extracted from the Twitter dataset.A simulated annealing algorithm searches for relevant features and selects and identifies sentimental terms that customers criticize.SA-MSVM is implemented,experimented with MATLAB,and the results are verified.The results concluded that SA-MSVM has more potential in sentiment analysis and classification than the existing Support Vector Machine(SVM)approach.SA-MSVM has obtained 96.34%accuracy in classifying the product review compared with the existing systems.

Chinese micro-blog sentiment classification through a novel hybrid learning model

With the rising and spreading of micro-blog, the sentiment classification of short texts has become a research hotspot. Some methods have been developed in the past decade. However, since the Chinese and English are different in language syntax, semantics and pragmatics, sentiment classification methods that are effective for English twitter may fail on Chinese micro-blog. In addition, the colloquialism and conciseness of short Chinese texts introduces additional challenges to sentiment classification. In this work, a novel hybrid learning model was proposed for sentiment classification of Chinese micro-blogs, which included two stages. In the first stage, emotional scores were calculated over the whole dataset by utilizing an improved Chinese-oriented sentiment dictionary classification method. Data with extremely high or low scores were directly labeled. In the second stage, the remaining data were labeled by using an integrated classification method based on sentiment dictionary, support vector machine(SVM) and k-nearest neighbor(KNN). An improved feature selection method was adopted to enhance the discriminative power of the selected features. The two-stage hybrid framework made the proposed method effective for sentiment classification of Chinese micro-blogs. Experiments on the COAE2014(Chinese Opinion Analysis Evaluation 2014) dataset show that the proposed method outperforms other schemes.

Automatic Sentimental Analysis by Firefly with Levy and Multilayer Perceptron

The field of sentiment analysis(SA)has grown in tandem with the aid of social networking platforms to exchange opinions and ideas.Many people share their views and ideas around the world through social media like Facebook and Twitter.The goal of opinion mining,commonly referred to as sentiment analysis,is to categorise and forecast a target’s opinion.Depending on if they provide a positive or negative perspective on a given topic,text documents or sentences can be classified.When compared to sentiment analysis,text categorization may appear to be a simple process,but number of challenges have prompted numerous studies in this area.A feature selection-based classification algorithm in conjunction with the firefly with levy and multilayer perceptron(MLP)techniques has been proposed as a way to automate sentiment analysis(SA).In this study,online product reviews can be enhanced by integrating classification and feature election.The firefly(FF)algorithm was used to extract features from online product reviews,and a multi-layer perceptron was used to classify sentiment(MLP).The experiment employs two datasets,and the results are assessed using a variety of criteria.On account of these tests,it is possible to conclude that the FFL-MLP algorithm has the better classification performance for Canon(98%accuracy)and iPod(99%accuracy).

Aspect-Level Sentiment Analysis Based on Deep Learning

In recent years,deep learning methods have developed rapidly and found application in many fields,including natural language processing.In the field of aspect-level sentiment analysis,deep learning methods can also greatly improve the performance of models.However,previous studies did not take into account the relationship between user feature extraction and contextual terms.To address this issue,we use data feature extraction and deep learning combined to develop an aspect-level sentiment analysis method.To be specific,we design user comment feature extraction(UCFE)to distill salient features from users’historical comments and transform them into representative user feature vectors.Then,the aspect-sentence graph convolutional neural network(ASGCN)is used to incorporate innovative techniques for calculating adjacency matrices;meanwhile,ASGCN emphasizes capturing nuanced semantics within relationships among aspect words and syntactic dependency types.Afterward,three embedding methods are devised to embed the user feature vector into the ASGCN model.The empirical validations verify the effectiveness of these models,consistently surpassing conventional benchmarks and reaffirming the indispensable role of deep learning in advancing sentiment analysis methodologies.

Sentiment Classification Based on Piecewise Pooling Convolutional Neural Network

Recently,the effectiveness of neural networks,especially convolutional neural networks,has been validated in the field of natural language processing,in which,sentiment classification for online reviews is an important and challenging task.Existing convolutional neural networks extract important features of sentences without local features or the feature sequence.Thus,these models do not perform well,especially for transition sentences.To this end,we propose a Piecewise Pooling Convolutional Neural Network(PPCNN)for sentiment classification.Firstly,with a sentence presented by word vectors,convolution operation is introduced to obtain the convolution feature map vectors.Secondly,these vectors are segmented according to the positions of transition words in sentences.Thirdly,the most significant feature of each local segment is extracted using max pooling mechanism,and then the different aspects of features can be extracted.Specifically,the relative sequence of these features is preserved.Finally,after processed by the dropout algorithm,the softmax classifier is trained for sentiment classification.Experimental results show that the proposed method PPCNN is effective and superior to other baseline methods,especially for datasets with transition sentences.

Sentiment Analysis on the Social Networks Using Stream Algorithms

The rising popularity of online social networks (OSNs), such as Twitter, Facebook, MySpace, and LinkedIn, in recent years has sparked great interest in sentiment analysis on their data. While many methods exist for identifying sentiment in OSNs such as communication pattern mining and classification based on emoticon and parts of speech, the majority of them utilize a suboptimal batch mode learning approach when analyzing a large amount of real time data. As an alternative we present a stream algorithm using Modified Balanced Winnow for sentiment analysis on OSNs. Tested on three real-world network datasets, the performance of our sentiment predictions is close to that of batch learning with the ability to detect important features dynamically for sentiment analysis in data streams. These top features reveal key words important to the analysis of sentiment.

Optimization of Sentiment Analysis Using Teaching-Learning Based Algorithm

Feature selection and sentiment analysis are two common studies that are currently being conducted;consistent with the advancements in computing and growing the use of social media.High dimensional or large feature sets is a key issue in sentiment analysis as it can decrease the accuracy of sentiment classification and make it difficult to obtain the optimal subset of the features.Furthermore,most reviews from social media carry a lot of noise and irrelevant information.Therefore,this study proposes a new text-feature selection method that uses a combination of rough set theory(RST)and teaching-learning based optimization(TLBO),which is known as RSTLBO.The framework to develop the proposed RSTLBO includes numerous stages:(1)acquiring the standard datasets(user reviews of six major U.S.airlines)which are used to validate search result feature selection methods,(2)preprocessing of the dataset using text processing methods.This involves applying text processing methods from natural language processing techniques,combined with linguistic processing techniques to produce high classification results,(3)employing the RSTLBO method,and(4)using the selected features from the previous process for sentiment classification using the Support Vector Machine(SVM)technique.Results show an improvement in sentiment analysis when combining natural language processing with linguistic processing for text processing.More importantly,the proposed RSTLBO feature selection algorithm is able to produce an improved sentiment analysis.

基于跨模态交叉注意力网络的多模态情感分析方法

挖掘不同模态内信息和模态间信息有助于提升多模态情感分析的性能,本文为此提出一种基于跨模态交叉注意力网络的多模态情感分析方法。首先,利用VGG-16网络将多模态数据映射到全局特征空间;同时,利用Swin Transformer网络将多模态数据映射到局部特征空间;其次,构造模态内自注意力和模态间交叉注意力特征;然后,设计一种跨模态交叉注意力融合模块实现不同模态内和模态间特征的深度融合,提升多模态特征表达的可靠性;最后,通过Softmax获得最终预测结果。在2个开源数据集CMU-MOSI和CMU-MSOEI上进行测试,本文模型在七分类任务上获得45.9%和54.1%的准确率,相比当前MCGMF模型,提升了0.66%和2.46%,综合性能提升显著。

基于视觉注意力的图文跨模态情感分析

针对单模态情感分析无法完全捕获情感信息的问题,提出一种图像和文本跨模态情感分析模型(BERT-VistaNet),该模型没有直接使用视觉信息作为特征,而是利用视觉信息作为对齐方式,使用注意力机制指出文本中重要的句子,得到基于视觉注意力的文档表示。对于视觉注意力无法完全覆盖的文本内容,使用BERT模型对文本进行情感分析,得到基于文本的文档表示,将特征进行融合应用于情感分类任务。在Yelp公开餐厅数据集上,该模型相比基线模型TFN-aVGG,准确率提高了43%,相比VistaNet模型准确率提高了1.4%。

融合多窗口特征的词对标记情感三元组抽取

方面情感三元组抽取旨在从句子中抽取方面词、意见词和对应的情感极性。针对目前研究未充分挖掘局部上下文语义信息,缺乏对局部范围内的方面意见词对关联学习,以及遭受错误传播等问题,提出一种融合多窗口特征的词对标记情感三元组抽取模型。该模型利用BERT对句子信息进行处理,获取句子编码特征,采用多窗口特征学习机制学习局部范围内的情感特征关联,并挖掘句子包含的潜在语义信息,使用多头注意力图转换模块将所学习到的特征聚合成标记分布概率,利用改进的词对标记方案标记句子并解码得到三元组。在SemEval-ASTE的四个基准数据集上进行实验分析,相比GTS-BERT模型,所提模型在三元组抽取任务上F1分值分别提高了2.33、6.57、2.97、4.84个百分点。实验结果表明,所提模型可以有效学习局部语义信息,准确标记方面意见跨度,较为精确地提取情感三元组。

基于CLIP和交叉注意力的多模态情感分析模型

针对多模态情感分析中存在的标注数据量少、模态间融合不充分以及信息冗余等问题,提出了一种基于对比语言-图片训练(CLIP)和交叉注意力(CA)的多模态情感分析(MSA)模型CLIP-CA-MSA。首先,该模型使用CLIP预训练的BERT模型、PIFT模型来提取视频特征向量与文本特征;其次,使用交叉注意力机制将图像特征向量和文本特征向量进行交互,以加强不同模态之间的信息传递;最后,利用不确定性损失特征融合后计算输出最终的情感分类结果。实验结果表明:该模型比其他多模态模型准确率提高5百分点至14百分点,F1值提高3百分点至12百分点,验证了该模型的优越性,并使用消融实验验证该模型各模块的有效性。该模型能够有效地利用多模态数据的互补性和相关性,同时利用不确定性损失来提高模型的鲁棒性和泛化能力。

基于跨模态情感联合增强网络的多模态情感分析方法

多模态情感分析是人工智能领域重要的研究方向之一,旨在利用多模态数据判断用户情感。现有的大多数方法忽略了不同模态数据之间的异质性,导致情感分析结果出现偏差。针对以上问题提出一种基于跨模态情感联合增强网络的多模态情感分析方法。首先,利用3种深度神经网络预训练模型提取不同模态的语义特征,并通过双向长短期记忆网络挖掘其单模态上下文时序信息;其次,设计了一种跨模态情感联合增强模块,实现融合文本模态和视觉模态特征生成情感极性语义特征,融合文本模态和音频模态信息生成情感强度语义特征,并以情感极性作为方向情感强度表示增幅联合增强情感语义。通过两个公共基准数据集CMU-MOSI和CMU-MOSEI的实验结果表明,所提出的跨模态情感联合增强网络可以获得比相关方法更好的性能。

融合Transformer和交互注意力网络的方面级情感分类模型

现有的大多数研究者使用循环神经网络与注意力机制相结合的方法进行方面级情感分类任务。然而,循环神经网络不能并行计算,并且模型在训练过程中会出现截断的反向传播、梯度消失和梯度爆炸等问题,传统的注意力机制可能会给句子中重要情感词分配较低的注意力权重。针对上述问题,该文提出了一种融合Transformer和交互注意力网络的方面级情感分类模型。首先利用BERT(bidirectional encoder representation from Transformers)预训练模型来构造词嵌入向量,然后使用Transformer编码器对输入的句子进行并行编码,接着使用上下文动态掩码和上下文动态权重机制来关注与特定方面词有重要语义关系的局部上下文信息。最后在5个英文数据集和4个中文评论数据集上的实验结果表明,该文所提模型在准确率和F1上均表现最优。

融合双通道的语义信息的方面级情感分析

针对方面级情感分析任务中语义信息难以提取以及方面词信息难以和上下文信息相关联的问题,提出一种融合双通道的语义信息模型(FDCS)。通过BERT预训练模型搭建两个通道获取不同层次的语义信息,一个是全局信息通道,另一个是句子信息通道;使用语义注意力融合双通道中不同层次的语义信息,将融合后的语义信息再次分别融入全局信息和句子信息;根据每个通道语义信息的不同分别提取相应的特征信息。在3个基准数据集上的实验结果表明,该模型的性能优于其它模型。

基于依赖类型剪枝的双特征自适应融合网络用于方面级情感分析

现有的模型将基于依赖树的图神经网络用于方面级情感分析,一定程度上提升了模型的分类性能。然而,由于依赖解析技术的限制,语法解析结果的不精确导致依赖树存在大量噪声,使得模型的性能提升有限。此外,一些句子本身并不符合标准的句法结构。以往的研究以同样的置信度利用句法信息和语义信息,没有充分考虑它们对于确定方面词极性的贡献的不同,导致模型在相应的数据集上性能较差。为了克服这些困难,文中提出了一种基于依赖类型剪枝的双特征自适应融合网络。具体来说,该模型使用一种新型的混合方法,命名为依赖关系类型剪枝和邻接矩阵平滑,来缓解句法解析产生的噪声。此外,该模型通过双特征自适应融合模块充分考虑句子的句法信息的可用程度,以一种更灵活的方式将句法特征和语义特征结合起来用于方面级情感分析。在5个公开可用的数据集上进行广泛的实验,结果证明了该方法明显优于基线模型。

基于KMeans-EDA算法的非均衡评论情感分类研究

学习者真实的评价是反映在线课程优缺点的重要指标,快速准确地获得其反馈,对于在线课程的优化极为重要。为深入挖掘学习者的在线学习行为,继而为在线教学提供有效的数据基础,爬取了中国大学MOOC平台的课程评论文本,基于Bert模型的结构,建立了基于自注意力文本表征的机器学习模型,能够实现对评论文本的精确情感分类,从而获得学习者内隐的情感状态。由于爬取数据中负面评论较少,故设计了KMeans-EDA自适应均衡采样训练策略,解决了训练过程中模型偏向多数类的问题,提升了模型对负面评论的识别能力。实验结果表明,该策略可以将模型对评论文本的F1-score值从0.6902提升到0.7399。